last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0185185185185186,
  "eval_steps": 50,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14814814814814814,
      "grad_norm": NaN,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.14814814814814814,
      "eval_loss": NaN,
      "eval_runtime": 1.4117,
      "eval_samples_per_second": 8.5,
      "eval_steps_per_second": 2.125,
      "step": 1
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": NaN,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": NaN,
      "learning_rate": 1.5e-06,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": NaN,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": NaN,
      "learning_rate": 2.5e-06,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": NaN,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 1.0555555555555556,
      "grad_norm": NaN,
      "learning_rate": 3.5e-06,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 1.2037037037037037,
      "grad_norm": NaN,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 1.3518518518518519,
      "grad_norm": NaN,
      "learning_rate": 4.5e-06,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 1.5,
      "grad_norm": NaN,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 1.6481481481481481,
      "grad_norm": NaN,
      "learning_rate": 4.8776412907378845e-06,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 1.7962962962962963,
      "grad_norm": NaN,
      "learning_rate": 4.522542485937369e-06,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": NaN,
      "learning_rate": 3.969463130731183e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 2.111111111111111,
      "grad_norm": NaN,
      "learning_rate": 3.272542485937369e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 2.259259259259259,
      "grad_norm": NaN,
      "learning_rate": 2.5e-06,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 2.4074074074074074,
      "grad_norm": NaN,
      "learning_rate": 1.7274575140626318e-06,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 2.5555555555555554,
      "grad_norm": NaN,
      "learning_rate": 1.0305368692688175e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 2.7037037037037037,
      "grad_norm": NaN,
      "learning_rate": 4.774575140626317e-07,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 2.851851851851852,
      "grad_norm": NaN,
      "learning_rate": 1.223587092621162e-07,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 3.0185185185185186,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.4506054221824e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}